[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122466-en":3,"doc-seo-122466-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122466,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Coupling Machine Learning Local Predictions with a Computational Fluid Dynamics Solver to Accelerate Transient Buoyant Plume Simulations - Paper","Data-driven methods show strong potential for accelerating computational fluid dynamics (CFD), yet surrogate machine-learning models struggle with physical consistency and real-world scalability. This study proposes a versatile hybrid CFD–machine-learning framework to accelerate long-term incompressible flow simulations without accuracy loss. A neural network is trained offline on simulated two-dimensional transient buoyant plume flows to predict cell-level temporal pressure-field changes in comparable scenarios. Cell-level predictions enable application to diverse geometries without retraining. Predicted pressures serve as initial values to speed pressure–velocity coupling, improving the Poisson-equation initial guess by 94% on average and reducing the first corrector iteration factor by about 3, depending on the iterative solver.","arXiv :2409 .07175v1 [physics .flu-dyn] 11 Sep 2024  \nTwelfth International Conference on  \nComputational Fluid Dynamics (ICCFD12),  \nKobe, Japan, July 14-19, 2024  \nCoupling Machine Learning Local Predictions with a Computational Fluid Dynamics Solver to Accelerate Transient Buoyant Plume Simulations  \nC. Caron∗ ,∗∗ , P. Lauret∗ and A. Bastide∗  \nCorresponding author: [clement.caron@univ-reunion.fr](clement.caron@univ-reunion.fr) ∗ Department of Sustainable Built Environment, PIMENT lab, University of Reunion, La Réunion, France.  \n∗∗ Research & Development team, INTEGRALE Ingénierie, Saint-Gilles les Hauts, La  \nRéunion, France.  \nAbstract:  \nData-driven methods demonstrate considerable potential for accelerating the inherently expensive computational fluid dynamics (CFD) solvers. Nevertheless, pure machine-learning surrogate models face challenges in ensuring physical consistency and scaling up to address real-world problems. This study presents a versatile and scalable hybrid methodology, combining CFD and machine learning, to accelerate long-term incompressible fluid flow simulations without compromising accuracy. A neural network was trained offline using simulated data of various two-dimensional transient buoyant plume flows. The objective was to leverage local features to predict the temporal changes in the pressure field in comparable scenarios. Due to cell-level predictions, the methodology was successfully applied to diverse geometries without additional training. Pressure estimates were employed as initial values to accelerate the pressure-velocity coupling procedure. The results demonstrated an average improvement of 94% in the initial guess for solving the Poisson equation. The first pressure corrector acceleration reached a mean factor of 3, depending on the iterative solver employed. Our work reveals that machine learning estimates at the cell level can enhance the efficiency of CFD iterative linear solvers while maintaining accuracy. Although the scalability of the methodology to more complex cases has yet to be demonstrated, this study underscores the prospective value of domain-specific hybrid solvers for CFD.  \nKeywords: Computational fluid dynamics, Machine learning, buoyant plume, incompressible flow, Poisson equation.  \n1 Introduction  \nComputational fluid dynamics (CFD) techniques enable the simulation of fluid flow, which is valuable for a wide range of scientific challenges. This flexible numerical approach can model various physical phenomena across different time and space scales based on the governing equations. As a result, CFD finds applications in many areas, such as aerodynamics, turbomachinery, hydrology, chemical processes, meteorology, and buildings [1, 2] . However, the computational time required to solve unsteady large-scale real-world problems remains a significant bottleneck. Despite the development of efficient numerical approaches and the growth of computational power, CFD is inaccessible for numerous applications [3, 4] . In this context, the CFD community is paying greater attention to machine learning algorithms to accelerate solvers by creating cost-effective models [3, 5 , 6 , 7 , 8] . The machine learning field encompasses various algorithms that extract valuable information from data, leading to applications such as pattern recognition or surrogate modeling [9] . Data-driven models can provide fast predictions, rendering thema desirable alternative to high-fidelity physics-based simulations. Nevertheless, further research is required to identify effective ways to combine CFD with machine learning to speed up simulations while maintaining accuracy [10] .  \nAlthough machine learning algorithms are not novel, they have substantially advanced recently. A combination of factors, including abundant data, more capable hardware, increased computational power, and the development of efficient algorithms, has revealed the potential of machine learning. Deep learning, which relies on deep","cbCaibDdtpxykWKh","https://ap.wps.com/l/cbCaibDdtpxykWKh","pdf",3015860,1,18,"English","en",105,"# Abstract\n# Introduction\n## CFD applications and computational bottlenecks\n## Machine learning for solver acceleration\n## Challenges and physics-aware hybrid approaches","[{\"question\":\"Why can pure machine-learning surrogates be difficult for CFD acceleration?\",\"answer\":\"They may fail to guarantee physical consistency and often face challenges in scaling to real-world problems while preserving accuracy.\"},{\"question\":\"How does the proposed hybrid method use machine learning with CFD?\",\"answer\":\"A neural network trained offline predicts cell-level temporal pressure-field changes, while CFD uses these pressure estimates as initial values for pressure–velocity coupling.\"},{\"question\":\"What performance improvements were reported for the Poisson equation and pressure corrector steps?\",\"answer\":\"The approach improves the initial guess for the Poisson equation by an average of 94%. The first pressure corrector acceleration achieved a mean factor of 3, depending on the iterative solver.\"}]","Coupling Machine Learning Local Predictions with a Computational Fluid Dynamics Solver to Accelerate Transient Buoyant Plume Simulations - Paper | PDF",1785810802,45,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"coupling-machine-learning-local-predictions-with-a-computational-fluid-dynamics-solver-to-accelerate-transient-buoyant-plume-simulations-paper","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/coupling-machine-learning-local-predictions-with-a-computational-fluid-dynamics-solver-to-accelerate-transient-buoyant-plume-simulations-paper/122466/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why can pure machine-learning surrogates be difficult for CFD acceleration?","Question",{"text":75,"@type":76},"They may fail to guarantee physical consistency and often face challenges in scaling to real-world problems while preserving accuracy.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed hybrid method use machine learning with CFD?",{"text":80,"@type":76},"A neural network trained offline predicts cell-level temporal pressure-field changes, while CFD uses these pressure estimates as initial values for pressure–velocity coupling.",{"name":82,"@type":73,"acceptedAnswer":83},"What performance improvements were reported for the Poisson equation and pressure corrector steps?",{"text":84,"@type":76},"The approach improves the initial guess for the Poisson equation by an average of 94%. The first pressure corrector acceleration achieved a mean factor of 3, depending on the iterative solver.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]